Detection of Small-Sized Electronics Endangering Facilities Involved in Recycling Processes Using Deep Learning

In Japan, local governments implore residents to remove the batteries from small-sized electronics before recycling them, but some products still contain lithium-ion batteries. These residual batteries may cause fires, resulting in serious injuries or property damage. Explosive materials such as mob...

Full description

Saved in:
Bibliographic Details
Main Authors: Zizhen Liu, Shunki Kasugaya, Nozomu Mishima
Format: Article
Language:English
Published: MDPI AG 2025-03-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/5/2835
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850228234884481024
author Zizhen Liu
Shunki Kasugaya
Nozomu Mishima
author_facet Zizhen Liu
Shunki Kasugaya
Nozomu Mishima
author_sort Zizhen Liu
collection DOAJ
description In Japan, local governments implore residents to remove the batteries from small-sized electronics before recycling them, but some products still contain lithium-ion batteries. These residual batteries may cause fires, resulting in serious injuries or property damage. Explosive materials such as mobile batteries (such as power banks) have been identified in fire investigations. Therefore, these fire-causing items should be detected and separated regardless of whether small-sized electronics recycling or other recycling processes are in use. This study focuses on the automatic detection of fire-causing items using deep learning in recycling small-sized electronic products. Mobile batteries were chosen as the first target of this approach. In this study, MATLAB R2024b was applied to construct the You Only Look Once version 4 deep learning algorithm. The model was trained to enable the detection of mobile batteries. The results show that the model’s average precision value reached 0.996. Then, the target was expanded to three categories of fire-causing items, including mobile batteries, heated tobacco (electronic cigarettes), and smartphones. Furthermore, real-time object detection on videos using the trained detector was carried out. The trained detector was able to detect all the target products accurately. In conclusion, deep learning technologies show significant promise as a method for safe and high-quality recycling.
format Article
id doaj-art-625b362aaf8d4586acbcf0c32f85abea
institution OA Journals
issn 2076-3417
language English
publishDate 2025-03-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-625b362aaf8d4586acbcf0c32f85abea2025-08-20T02:04:35ZengMDPI AGApplied Sciences2076-34172025-03-01155283510.3390/app15052835Detection of Small-Sized Electronics Endangering Facilities Involved in Recycling Processes Using Deep LearningZizhen Liu0Shunki Kasugaya1Nozomu Mishima2Graduate School of Engineering Science, Akita University, Akita 010-8502, JapanGraduate School of Engineering Science, Akita University, Akita 010-8502, JapanGraduate School of Engineering Science, Akita University, Akita 010-8502, JapanIn Japan, local governments implore residents to remove the batteries from small-sized electronics before recycling them, but some products still contain lithium-ion batteries. These residual batteries may cause fires, resulting in serious injuries or property damage. Explosive materials such as mobile batteries (such as power banks) have been identified in fire investigations. Therefore, these fire-causing items should be detected and separated regardless of whether small-sized electronics recycling or other recycling processes are in use. This study focuses on the automatic detection of fire-causing items using deep learning in recycling small-sized electronic products. Mobile batteries were chosen as the first target of this approach. In this study, MATLAB R2024b was applied to construct the You Only Look Once version 4 deep learning algorithm. The model was trained to enable the detection of mobile batteries. The results show that the model’s average precision value reached 0.996. Then, the target was expanded to three categories of fire-causing items, including mobile batteries, heated tobacco (electronic cigarettes), and smartphones. Furthermore, real-time object detection on videos using the trained detector was carried out. The trained detector was able to detect all the target products accurately. In conclusion, deep learning technologies show significant promise as a method for safe and high-quality recycling.https://www.mdpi.com/2076-3417/15/5/2835YOLOv4object detectionrecyclingsmall-sized electronicsfire prevention
spellingShingle Zizhen Liu
Shunki Kasugaya
Nozomu Mishima
Detection of Small-Sized Electronics Endangering Facilities Involved in Recycling Processes Using Deep Learning
Applied Sciences
YOLOv4
object detection
recycling
small-sized electronics
fire prevention
title Detection of Small-Sized Electronics Endangering Facilities Involved in Recycling Processes Using Deep Learning
title_full Detection of Small-Sized Electronics Endangering Facilities Involved in Recycling Processes Using Deep Learning
title_fullStr Detection of Small-Sized Electronics Endangering Facilities Involved in Recycling Processes Using Deep Learning
title_full_unstemmed Detection of Small-Sized Electronics Endangering Facilities Involved in Recycling Processes Using Deep Learning
title_short Detection of Small-Sized Electronics Endangering Facilities Involved in Recycling Processes Using Deep Learning
title_sort detection of small sized electronics endangering facilities involved in recycling processes using deep learning
topic YOLOv4
object detection
recycling
small-sized electronics
fire prevention
url https://www.mdpi.com/2076-3417/15/5/2835
work_keys_str_mv AT zizhenliu detectionofsmallsizedelectronicsendangeringfacilitiesinvolvedinrecyclingprocessesusingdeeplearning
AT shunkikasugaya detectionofsmallsizedelectronicsendangeringfacilitiesinvolvedinrecyclingprocessesusingdeeplearning
AT nozomumishima detectionofsmallsizedelectronicsendangeringfacilitiesinvolvedinrecyclingprocessesusingdeeplearning